Lifting potential growth in the euro area
Speech by Peter Praet, Member of the Executive Board of the ECB,
at the Welt-Währungskonferenz,
Berlin, 23 April 2015
Ladies and gentlemen ,
It is a pleasure to be here today and to have an opportunity to address this gathering.
As you know, over the last two years the ECB has taken a series of monetary policy measures to ensure that we meet our price stability mandate. In my remarks today I would like to look back on those measures and the effects they have had, but to do so in a broader context – that is, by looking at overall growth trends within the euro area and how monetary policy fits within them.
The main point I would like to make is that the euro area needs a combination of policies for the cyclical recovery we are seeing to become a lasting one: accommodative monetary policy and determined structural reforms need to go hand-in-hand.
Potential growth and monetary policy
The euro area economy seems now to be turning the corner. Both the hard and soft data suggest that the activity is gathering momentum and looks set to strengthen over the course of this year. Consistent with this, all the major forecasting institutions have revised up their expectations for GDP growth in 2015 and the years ahead. We are therefore seeing the beginnings of a cyclical recovery. But it is not yet a structural one.
What I mean by this is that, though the business cycle is improving, the notable decline in euro area’s potential growth rate has not been addressed, which can be imagined as the “speed limit” of the economy – the rate at which it can grow while maintaining stable inflation. International institutions currently estimate the potential growth rate to be below 1% in the euro area, compared with above 2% in the United States. 
This fall in potential growth is not a new phenomenon. Potential growth in the euro area has been declining since the late 1990s, driven mostly by a prolonged slowdown in total factor productivity growth, a trend decline in hours worked and the labour participation rate growing more hesitantly. Before the crisis, however, it seems that this deceleration went largely unnoticed. And this has in fact contributed both to the type of crisis we have experienced and the situation of weak potential growth that we face now.
In several countries expectations of future income became detached from these underlying trends in potential growth in the pre-crisis period, leading the public and private sectors in different countries to accumulate excessive debt. This in turn left parts of the euro area facing not a normal business cycle, but what is known as a “balance sheet recession”, where the recovery is hampered by the need for governments, households and firms to deleverage. And so this also creates financial fragmentation.
The difficulty of managing this special type of crisis, coupled with what were in hindsight some errors in policy sequencing, then noticeably exacerbated the fall in potential growth. Both the level and growth rate of potential fell steeply after 2008 and have barely recovered since.
This is due to, first, a deep and protracted decline in investment – nearly 20% from peak to trough – which has caused the capital stock to grow more slowly and increased its average age. Seen in historical perspective, this deceleration since 2008 has been substantially stronger than in previous recessions. Of particular concern is the growth rate of the machinery and equipment stock, the most productive component, which has slowed the most.
Second, long spells of crisis-related unemployment have resulted in so-called “hysteresis effects” in the labour market – that is, where workers are unemployed for so long that they lose their skills and become unemployable. As result, structural unemployment (unemployment caused by a mismatch between jobs and skills) has risen in the euro area as whole, increasing from just under 9% in 2008 to almost 10% in 2014 , and much more in some stressed countries.
Low potential growth matters for several reasons. It creates a vicious circle where firms and households hold back consumption and investment due to diminished expectations, which in turn lowers potential growth further. It makes it harder for the economy as a whole to grow out of debt. And, if sustained, it implies a permanent loss in productive capacity – and hence in jobs – which has social consequences. Indeed, if potential growth in the euro area had carried on rising at the same rate as in 2007, potential output would currently be more than 4% higher.
But low potential growth also has important implications for monetary policy. Let me highlight two that are particularly significant.
First, to ensure medium-term price stability, monetary policy aims to limit how much actual growth diverges from potential growth. In the textbook representation it is assumed that the monetary policy stance is calibrated to close the gap between activity and potential, the so-called “output gap”. Therefore, if potential changes, the output gap can change as well, and the appropriate stance of monetary policy has to be adjusted. But exactly how falling potential growth affects monetary policy is not straightforward.
At first sight, one would think that it would make achieving price stability easier as, for a given level of output, the output gap that monetary policy has to close is smaller and inflation would return to the central bank’s objective more quickly. Yet this would be a Pyrrhic victory, as it implies that the central bank would have to reign in its accommodative policy while there are still very high levels of structural unemployment. And in any event, the reality is more complex.
As I already intimated, if potential growth falls, expectations of incomes and profits fall as well, and firms and households are likely to consume and invest less today. Such a feedback from potential growth to current growth would in fact have the opposite effect – it would prevent the output gap from closing – and create downward pressure on inflation. In other words, falling potential would not make the task of monetary policy easier, but harder.
Second, to steer output back towards potential, monetary policy aims to bring its policy interest rates as close as possible to what is known as the “equilibrium” interest rate – that is, the interest rate where resources are fully employed in the economy and inflation is stable.
A lower rate of potential growth can however reduce the equilibrium interest rate, as for example lower growth in either productivity or labour supply reduces demand for investment. And this can in turn create constraints for monetary policy, since in order to “shadow” the equilibrium rate the central bank may have to reduce its policy rates to zero or lower. Moreover, it may have keep rates at those levels more often in the future when the economy slumps and inflation deviates from its objective.
It is difficult to quantify exactly these relationships as they involve unobservable concepts such as the output gap and the equilibrium interest rate, which are often very difficult to estimate in real-time and subject to revisions. But the outcomes we can observe suggest that falling potential growth has been accompanied by a widening not a closing output gap, with inflation following a prolonged downward drift, falling by around 3.6 percentage points between December 2011 and January 2015.
And – consistent with a lower equilibrium rate – monetary policy has found itself constrained: interest rates have had to go down to very low levels in order to bring inflation in line with our aim of below but close to 2% over the medium-term.
Clearly, both supply and demand policies have a role to play in addressing this situation. Supply policies to raise potential growth; and demand policies, including monetary policy, to close the output gap. But while both are necessary, monetary policy cannot wait for other policies to do their job. Regardless of the causes of low inflation – including the possible role of falling potential – monetary policy has to respond to the risks to the inflation outlook and meet its price stability mandate.
The ECB’s monetary policy response
So how have we responded to those disinflationary risks?
First of all, it is important to understand the initial conditions we faced when inflation began that downward drift. The debt overhang in the public sector meant the ability of fiscal policy to support demand was constrained. Nominal interest rates were by that point already low, limiting the scope for conventional monetary policy. And there was significant heterogeneity in financial conditions within the euro area, caused by the fragmentation I already described, which affected the transmission of monetary policy across countries.
We therefore used fully the remaining margin for nominal rate cuts, which meant cutting our main refinancing rate from 125 basis points in December 2011 to 5 basis points in September 2014. But faced with the possibility that the equilibrium interest rate might have fallen below that level, we have also taken steps to increase the traction of our low rates over the path of the economy. In particular, we have introduced two measures to strengthen the link between our short-term interest rates and longer-term rates, which as you know are the rates that matter most for the real economy.
First, we have provided forward guidance on our future policy, which has helped lower long-term rates by reducing uncertainty about the expected path of short-term rates. Second, we have introduced a negative interest rate on our deposit facility – the rate that banks pay when they deposit money at the ECB – which has led to a further flattening of the yield curve through its effect on expectations. 
Together, these policies have helped reduce market rates to very low levels and extend that impact out along the yield curve to longer borrowing maturities, where they have most effect on consumption and – in particular – investment.
I put this stress on investment because, in our current circumstances, it is particularly important for monetary policy. Investment increases demand today, which helps close the output gap and restore price stability. And it increases supply tomorrow, thus raising potential growth and allowing the equilibrium interest rate to rise. Remember also that the investment shortfall during the crisis is at the core of the recent deterioration of our potential growth rate.
What ultimately matters for investment, however, is not market interest rates but the cost of capital –how much long-term interest rates feed through into in firms’ and households’ cost of borrowing. And this is where heterogeneity matters. As late as the middle of last year viable firms were still struggling to access finance in parts of the euro area. The percentage of financially constrained but viable SMEs  is estimated to have varied from around 1% in Germany and Austria to a quarter of the SME population in Spain and as much as a third in Portugal. 
In this context, between June and September last year we launched our credit easing package with the aim to remove these transmission blockages. It consisted of, first, our targeted long-term refinancing operations, which provide cheap long-term funding to banks on the condition that they expand loans to the real economy. And second, our purchases of asset-backed securities and covered bonds, which have a relatively tight link with credit extension to the private sector.
At the same time, these measures were launched as the ECB’s Comprehensive Assessment of bank balance sheets was reaching its completion, which had led to banks frontloading their deleveraging and strengthening their balance sheets by over €200 billion in advance of the outcome. The new monetary policy impulse coming from these measures therefore arrived just as banks were in a stronger position to transmit it.
However, as the credit easing package was being rolled out in the second half of last year, the inflation situation materially worsened in the euro area. And while the main driver was the sharp drop in global oil prices around that time, there were strong indications that this might not be a solely transitory development.
Our analysis showed that the persistence of low inflation across a range of statistical metrics was very high. Inflation expectations had become more sensitive to that low inflation and were declining at all horizons. And measures of core inflation – without food and energy – had become less sticky, implying a higher risk that the oil price fall would feed into lower wages (so-called “second round effects”), and so would not just bounce back when that effect faded away.
This posed a risk to our mandate, and also the credit easing we were aiming to engineer. Remember that if second round effects set in and people start to expect lower inflation in the future, then real interest rates rise – that is, interest rates adjusted for inflation. And if this happens, then any easing of nominal credit conditions is effectively neutralised as the real cost of finance increases.
In fact, by January 2015 expected real rates had started rising – increasing by almost half a percentage point in the previous few months alone. The cost of deflation protection had also gone up by 185 basis points between 2 December 2014 and 9 January 2015, showing that investors saw a material risk that inflation could fall further and real interest rates could rise more.
In this situation, it was absolutely essential that we re-anchored inflation expectations and warded off this potential tightening. And as we could no longer cut our main interest rates, we resorted to asset purchases to achieve this. What was crucial for us was to send a very strong signal that we would not allow price stability to be jeopardised. The only asset class that we could purchase in large enough volume to have that effect was public sector securities.
The data so far suggests that this signal came at the right time and, as a result, our credit easing package is indeed having positive results.
Since we announced our purchase programme in January, real interest rates have fallen along the yield curve, driven by both a flattening of the term structure of interest rates and, crucially, by a reversal in the decline of inflation expectations. Market-based measures of inflation expectations have increased at all horizons, and as they rise back to levels consistent with medium-term price stability real interest rates should further decrease.
Bank lending data have also been steadily improving. The cost of borrowing for euro area firms has fallen by around 45 basis points since the summer of last year, with rates converging across euro area countries. Our most recent Bank Lending Survey shows that euro area banks are continuing to ease credit conditions.  Importantly, this is mainly being driven by an improved pass through from banks’ cost of funding to their lending rates in a context of higher competition for customers – which was exactly the reaction that our credit measures aimed to provoke.
At the same time, confidence in the recovery is improving considerably. The latest European Commission economic sentiment indicator improved notably in March and was driven by a pick-up in both consumer and business confidence. This improvement was also echoed in the latest data in Germany, with consumer confidence rebounding strongly in March and business confidence indicators continuing to expand.
All in all, this is creating a propitious environment for firms to undertake new investment. The more the real cost of capital falls, the greater the number of investment projects that will break even. And the more the recovery picks up, the greater the confidence firms will have to take such risky decisions.
This also applies to firms in this country. Although private investment is now strengthening, it has been consistently below expectations in recent years, and recent increases are mainly due to the construction sector. All the conditions are now in place to take advantage of the current environment and for business investment to grow more intensely.
A stronger recovery here driven by rising investment would not only benefit Germany, but also have positive spillovers to the rest of the euro area, then feeding back into German exports. According to ECB staff estimates, a 1% permanent increase in German real GDP would lead to an increase of around 0.3% in real GDP in France and Italy and a 0.2% increase in Spain. 
And ultimately, it is only once we have stronger euro area recovery with higher investment that potential growth will start to recover, which will in turn create the conditions for a rising equilibrium interest rate and a return to more normal levels of interest rates.
Lifting potential growth
Taking these improvements into account, the ECB is currently projecting the output gap to close and inflation to return close to 2% around 2017. This is expected to be driven by a rise in actual growth towards potential, rather than the other way around. Yet this does not make the challenge of raising potential growth any less important – even on this improved trajectory the output gap will still close with high levels of structural unemployment, which would not be socially acceptable.
Monetary policy can play a role in supporting potential growth through the investment channel I have already described. Moreover, if it is true that weak demand has reduced potential growth through hysteresis effects in the labour market, it could also be true that an acceleration in demand will support potential through the same mechanism working symmetrically – for example, as the labour market tightens firms may start hiring discouraged workers and retraining them.
Over the longer-term, however, monetary policy cannot increase growth. The growth rate of the economy is ultimately driven by factors such as the efficiency with which resources are allocated, the strength of incentives to innovate and invest, the ease of doing business, and the quality of public institutions – in other words, by the supply-side. And making improvements in all of these areas comes down chiefly to structural policies.
Importantly, our monetary policy is creating a unique window of opportunity to enact such reforms. When demand is weak structural reforms can result in higher short-term adjustment costs, as for example people losing their jobs take longer to find new ones. Financial frictions can also reduce the short-term confidence effects of reforms, as people cannot rely on consumption smoothing through credit to manage labour market risks. These effects can make introducing reforms harder.
With our expansionary monetary policy, however, we are supporting strengthening demand and better access to credit, and thus limiting the potential downsides of reforms in the short run. For this reason, there is every reason for national authorities to push ahead with structural reforms today.
There are many channels through which structural reforms can raise potential. For example, in some countries active labour market policies aimed at retraining vulnerable groups such as the low skilled, the long-term unemployed and the inactive can play a key role in getting people back to work – which is particularly important given the drag that ageing societies will create on labour supply. Another example is product market reforms that enhance competition in protected sectors, which can drive productivity gains by encouraging new innovative firms to enter the market and existing ones to improve their internal efficiency – a finding recently confirmed by the IMF. 
Making use of all these channels is necessary to lift growth potential. What is particularly important for the euro area today, however, is that we focus our efforts on reforms with the greatest short-term impact on demand and the lowest potential for disinflationary effects. Here there are two areas that I see as especially crucial, not least as they would directly complement our monetary policy by putting firms in a better position to exploit our low interest rates and invest in productive projects.
The first area is policies that help reduce the private debt overhang, which still persists in parts of the euro area. This weighs on both the recovery in credit supply – through a high stock of non-performing loans (NPLs) – and on higher credit demand, through constraining new investment.
The completion of the ECB’s Comprehensive Assessment has created the preconditions to address this problem, since many banks have now acknowledged NPLs and raised capital, putting them in a stronger position to begin restructuring loans to distressed borrowers. But the right structural conditions are also critical to make private debt workouts successful, such raising the efficiency of insolvency regimes, out-of-court restructuring frameworks and judicial systems.  For example, to resolve insolvency in Italy takes 1.8 years, compared with just 0.4 years in Ireland. 
The second key area is policies that boost productivity growth.
From 2000-14 total factor productivity has grown by just 1.4% in the euro area, while in the US it has risen by 10.7%. Were productivity growth in the euro area to remain at such low levels, the set of profitable investment projects, even with very low long-run real rates, would not expand significantly. In other words, raising productivity would make our monetary policy commensurately more powerful.
Broadly speaking, total factor productivity can be raised either by firms innovating and adopting new technology, or by reallocating resources within and across sectors towards the most productive firms. Often these are seen as separate issues, the first linked to human capital and R&D and only responding to public policy with long lags (e.g. through education); the second linked to product and labour market conditions and more directly responsive to government initiatives. But in fact product and labour market reforms can lift productivity through both channels.
For example, structural reforms that remove thresholds on firm growth – such as the regulations that kick-in in France when a firm exceeds 50 employees – can aid reallocation by accelerating the “up or out” process of firm turnover. But they can also improve technology adoption, because firm smallness and lack of ICT diffusion tend to be related. Italy, for example, has some of the lowest integration of digital technology in Europe, and an important explanation for this is the predominance of micro and small firms in the economy. 
Recent firm-level analysis by the Eurosystem suggests there is significant scope for policies aimed at encouraging firm growth and at improving reallocation to lift productivity, as within individual euro area countries there are a few highly productive firms and many with low productivity. And some recent evidence shows that – following the reforms undertaken during crisis in vulnerable countries – credit and labour are now being reallocated towards more productive uses. 
Progress in terms of technology diffusion will however likely take more time and further reforms to become visible, as it tends to require – on the government side – labour market reforms that allow firms to reorganise internally and optimise that technology, and – on the firm side – parallel investment in intangible capital such as management systems and organisational processes.
So I think we can view the euro area’s weak productivity performance in recent times as also an opportunity. Since many member states are far from the frontier of best practice, the potential magnitude of productivity gains is greater. To give an illustration of what is possible, simulations by researchers at the OECD suggest that a broad package of labour, product, tax and pension reforms would raise GDP per capita by about 11% after ten years for the average EU country under relatively quick reform implementation. The equivalent for the US is under 5%. 
Let me conclude.
What I have aimed to show today is that a lasting exit from the crisis requires a combination of policies. It requires monetary policy to do its job of delivering price stability; but it also requires structural policies to tackle decisively the euro area’s weak growth potential.
Raising growth potential is necessary for its own sake – to bring down structural unemployment and increase welfare. But is also helps monetary policy by supporting demand today, and by making it less likely that we run up against zero interest rates again in the future.
The incoming data suggest that the ECB’s measures are now gaining traction. The recovery is firming, and inflation expectations seem to be moving towards values more consistent with our aim. But it is only through actions by other stakeholders – governments, social partners and firms – that this cyclical improvement can be made permanent.
I would like to thank John Hutchinson for his contributions to this speech.
European Commission, Winter 2015 projections; IMF, April 2015 WEO; OECD, November 2014 Economic Outlook.
Average of IMF, Commission and OECD estimates.
For a fuller explanation of this point see speech by Peter Praet on “Price stability: a sinking will-o'-the-wisp?”, Washington, 16 April 2015.
Defined as those with positive turnover in the last six months seeking a bank loan.
See speech by Mario Draghi, “Monetary policy in prolonged period of low inflation”, Sintra, 26 May 2014.
ECB Bank Lending Survey, April 2015.
“Competitiveness and external imbalances within the euro area”, GVAR Handbook Structure and Applications of a Macro Model of the Global Economy for Policy Analysis, 2013.
IMF World Economic Outlook, April 2015, Box 3.5.
For more information on this point see speech by Peter Praet, “Repairing the bank lending channel: the next steps”, London, 17 November 2014.
Doing Business 2014: Understanding Regulations for Small and Medium-Size Enterprises, World Bank.
Around 90% of the firm population in Italy are firms with under 10 employees.
Bartelsman, E., F. di Mauro and E. Dorrucci (2015),“Eurozone rebalancing: Are we on the right track for growth? Insights from the CompNet micro-based data”, www.voxeu.org.
Bouis, R and R Duval (2011), “Raising the Potential Growth after the Crisis: A Quantitative Assessment of the Potential Gains from Various Structural Reforms in the OECD Area and Beyond”, OECD Economics Department Working Paper No. 835.